ABSTRACT

The development of a potential pharmaceutical requires establishing a risk prole that involves animal testing before human exposure. A pathologist is responsible for evaluating a representative list of tissues from study animals during the long, comprehensive toxicology screening process. During this review, the pathologist is often confronted with ndings that are of uncertain relationship to the test substance. The comparison of ndings in treated animal to those of the concurrent control group provides the initial (and most important) screen in order to make an assessment of a possible relationship to the test substance. There are those instances, however, wherein the concurrent control group will not contain sufcient numbers of animals to clearly make such an assessment. This is particularly true in toxicity studies with dogs and nonhuman primates wherein low numbers of animals are used, typically only 3/sex/group. Pathologists are often confronted with cases in which a nding in the high-dose group represents a tissue change that can represent an incidental nding, which might also be found in control group animals. In this example, as a result of biological variability, the incidence in the high-dose group might be elevated while the incidence in the control group is low or zero, implying a test substance relationship. A limited number of animals, again especially in dog and primate studies, make it difcult to dismiss a nding as unrelated to test substance unless there is a robust local historical control database that can be used to put the data into perspective. Lastly, for subtle changes, it is important to use reliable historical control data to make a proper assessment of potential compound-related effects (Dixon et al. 1995).